“Communication is the essence of life- it is how we acquire knowledge, attain goals, and form social bonds” -Unknown
Several years ago, Dr. Rupal Patel came across a young girl and an adult man having a conversation using their speech synthesizers; both using the same voice.
Patel said, “We wouldn’t give a little girl a prosthetic limb of a grown man. So why give her a grown man’s voice?”
And VocaliD was born.
Who is Rupal Patel?
Rupal Patel is a speech scientist and founder of VocaliD. She is a tenured professor at Northeastern University in Boston with joint appointments in the Department of Communication Sciences and Disorders and the College of Computer and Information Science.
She also founded and directs the Communication Analysis and Design Laboratory (CADLab). She is conducting research in the areas of the acquisition and impairment of speech prosody in healthy speakers and those with neuromotor disorders and the design of speech enhancement and learning technologies. This research and its production can change the lives of millions with speech disorders.
Giving Voice To 2.5 Million People
Each one of us has a unique voiceprint that reflects our age, size, gender, lifestyle, and even our personality. In the United States, there are 2.5 million people who are unable to speak due to developmental or medical conditions, and many use speech synthesizers on computerized devices to do so. This allows them to select the words they want to use and the device will then speak the words aloud. However, these devices only have a handful of options for voice output. Therefore, these voices are somewhat generic.
How VocaliD Began
Several years ago, Dr. Patel came across a young girl and an adult man having a conversation using their speech synthesizers. They were both using the same voice. Patel said, “We wouldn’t give a little girl a prosthetic limb of a grown man. So why give her a grown man’s voice?” Longfellow said “The human voice is the organ of the soul.”
So why must millions of individuals use just a limited number of voices when using a computerized device if our voice is unique?
Dr. Patel has set out to give individuals who are unable to speak their own unique voice.
VocaliD Is As Unique As a Fingerprint
VocaliD combines the speech of two individuals– the source and a surrogate filter. Prosodic features (intonation, stress, and rhythm) are preserved in a speaker even with speech disorders.
These sounds are recorded, even if it is just vowel sounds, to capture the prosodic features. Then, a volunteer voice donor records several hundred to several thousand sentences in order to capture all sounds and sound combinations that occur in the language.
Using technology, the prosodic features of the source and the recordings of the surrogate are combined to create a huge database for the text to speech software. It’s the blending of the two that leads to a unique voice for the recipient who now has a vocal identity unlike any other.
If you are interested in becoming a voice donor, you can go to https://www.vocalid.co/about.
Watch Rupal Patel speak at TEDWomen, 2013
Rupal Patel: Synthetic voices, as unique as fingerprints
Filmed December 2013 at TEDWomen 2013